diff options
author | Zhigang Gong <gongzg@freedesktop.org> | 2014-12-04 23:11:33 -0800 |
---|---|---|
committer | Zhigang Gong <gongzg@freedesktop.org> | 2014-12-04 23:11:33 -0800 |
commit | 715e11548c068a2ec8a1b5d69f1a439fa2696a6f (patch) | |
tree | 45efe48e1488d14e0d820f3e64275b1f06e34314 /Software/Beignet | |
parent | 4a7797562f2f3a494c4c5f65fdf8720c62582401 (diff) |
tewak formats.
Signed-off-by: Zhigang Gong <gongzg@freedesktop.org>
Diffstat (limited to 'Software/Beignet')
-rw-r--r-- | Software/Beignet/optimization-guide.mdwn | 6 |
1 files changed, 3 insertions, 3 deletions
diff --git a/Software/Beignet/optimization-guide.mdwn b/Software/Beignet/optimization-guide.mdwn index a223b661..b92814dc 100644 --- a/Software/Beignet/optimization-guide.mdwn +++ b/Software/Beignet/optimization-guide.mdwn @@ -32,7 +32,7 @@ size. We have the following rules: If use barrier, work group size should be equal or larger than 16 on baytrail and be equal or larger than 64 on other platforms. - If use SLM, here are a table: + If use SLM, here are a table: | Amount of SLM | 0 | 4K | 8K | 16K | 32K | | WorkGroup size| 16 | 64 | 128 | 256 | 512 | @@ -58,7 +58,7 @@ Memory Access Optimizations Currently, private buffer access in beignet backend is very slow. Many small private buffer could be optimized by the compiler. But the following type of dynamic indexed private buffer could not be optimized: - A bad example: + A bad example: ``` ... uint private_buffer[32]; @@ -69,7 +69,7 @@ Memory Access Optimizations } ``` - The following example is Ok for most cases: + The following example is Ok for most cases: ``` ... uint private_buffer[32]; |